Inorganic Chemistry, Vol. 16, No. 2, 1977 421
Tris(tripheny1 phosphite)cobalt(I) Halides 51 194-40-2;(BU~P)CO(DH)~NO~, 51 194-37-7; (Bu~P)CO(DH)~B~, (Bu~P)CO(DH)~C~, 24501-27-7; (Bu~P)CO(DH)~NCS, 51 194-38-8; (Bu~P)CO(DH)~CSH~NS, 61075-77-2; (BU,P)CO(DH)~C~H~N~O,(10) 5 1194-39-9; (Bu~P)CO(DH)~NO~, 61024-89-3; (Bu~P)CO(DH)~N~, 51 194-42-4; (Bu~P)CO(DH)CN,51 194-44-6; (Bu3P)Co(11) (Bu~P)CO(DH)~P(O)(OCH~)~, ( D H ) ~ S O Z C ~ H ~55886-56-1; CH~, (12) 61024-90-6; (Bu3P)Co(DH),CH2Br, 55886-62-9; (Bu3P)Co(DH)2C6H40CH3, 55886-64-1; (Bu~P)CO(DH)~C~HS, 55923-88-1; (Bu~P)CO(DH)~CH~, (13) ( B U ~ P ) C O ( D H ) ~ C ~ 55886-65-2; H~B~, 5 1020-41-8; [ (BU~P)ZCO(DH)~]AS(C~HS)~, 61026-07-1; Bupy, (14) 3978-81-2; (CH,O),P, 121-45-9; Bu~P,998-40-3; 13C,14762-74-4. Supplementary Material Available: Tables of 13CNMR shifts and P-C coupling constants for series I1 and 111, for additional compounds in series I, and for the X ligands, a more extensive tabulation similar to Table I1 but for ‘H NMR and 13CNMR for all three series, and a summary of the least-squares correlations (8 pages). Ordering information is given on any current masthead page.
(15)
References and Notes
(18)
K. D. Gailey, K. Igi, and B. E. Douglas, Inorg. Chem., 14,2956 (1975); P. L. Gaus and A. L. Crumbliss, ibid., 15, 739 (1976). T. B. Brill and A. J. Kotlar, Inorg. Chem., 13,470 (1974); J. M. Malin, C. F. Schmidt, and H. E. Toma, ibid., 14, 2924 (1975). K. W. Jennette, S. J. Lippard, and D. A. Ucko, Biochim. Biophys. Acta, 402,403 (1975). L. G. Marzilli, P. Politzer, W. C. Trogler, and R. C. Stewart, Inorg. Chem., 14, 2389 (1975). B. E. Mann, Chem. Commun., 976 (1971); L. F. Farnell, E. W. Randall, and E. Rosenberg, ibid., 1078 (1971); M. H. Chisholm, H. C. Clark, L. E. Manzer, and J. B. Stothers, J . Am. Chem. SOC.,94,5087 (1972); H. C. Clark and J. E. H. Ward. ibid.. 96, 1741 (1974). W. C. Trogler, R. C. Stewart, L. A. Epps, and Ll G. Marzilli, Znorg. Chem., 13, 1564 (1974).
(16) (17)
(19) (20) (21) (22) (23) (24) (25) (26) (27)
W. C. Trogler and L. G. Marzilli, J . Am. Chem. SOC.,96,7589 (1974). W. C. Trogler and L. G. Marzilli, Znorg. Chem., 14, 2942 (1975). H. M. McConnell, J . Chem. Phys., 27, 226 (1957). A. Carrington and A. D. McLachlan, “Introduction to Magnetic
Resonance”, Harper and Row, New York, N.Y., 1967. L. G. Marzilli, J. G. Salerno, and L. A. Epps, Inorg. Chem., 11, 2050 (1 972). L. GrMarzilli, R. C. Stewart, L. A. Epps, and J. B. Allen, J. Am. Chem. SOC.,95, 5796 (1973). J. B. Stothers, “Carbon-13 Nuclear Magnetic Resonance Spectroscopy”, Academic Press, New York, N.Y., 1972, pp 159, 250. See supplementary material. More information and figures may be found in the Ph.D. dissertation of R. C. Stewart, Jr., The Johns Hopkins University, 1976. (a) W . McFarlane, Proc. R . SOC.London, Ser. A , 306, 185 (1968); (b) F. J. Weigert and J. D. Roberts, J . Am. Chem. SOC.,91, 4940 (1969); (c) see ref 13, pp 376, 378. Theoretical studies on six-membered aromatic nitrogen heterocycles are reviewed in Chapter 7 of ref 13. L. G. Marzilli, L. A. Epps, T. Sorrell, and T. J. Kistenmacher, J . Am. Chem. SOC.,97, 3351 (1975). A. Bigotto, E. Zangrando, and L. Randaccio, J. Chem.Soc., Dalton Trans., 96 (1976). C. K. Johnson, “ORTEP, Report ORNL-3794, Oak Ridge National Laboratory, Oak Ridge, Tenn. M. R. Churchill, Znorg. Chem., 12, 1213 (1973). See ref 13, p 199. Similar relationships between ”C and ‘H NMR shifts were observed in Pt(I1) complexes: M. A. M. Meester, D. J. Stukens, and K. Vrieze, Inorg. Chim. Acta, 15, 137 (1975). T.G. Appleton, H. C. Clark, and L. E. Manzer, Coord. Chem. Rev., 10, 335 (1973). G. Singh, J. Organomet. Chem., 99, 251 (1975). G. Singh and G. S. Reddy, J . Organomet. Chem., 42, 267 (1972). K. R. Dixon, K. C. Moss, and M. Smith, J . Chern.SOC.,Dalton Trans., 990 (1974). W. McFarlane, J. Chem. SOC.,Dalton Trans., 324 (1974). D. R. Coulson, J . Am. Chem. Soc., 98, 31 11 (1976).
Contribution No. 2398 from the Central Research & Development Department, Experimental Station, E. I. du Pont de Nemours and Company, Wilmington, Delaware 19898
Synthesis and Properties of Cobalt(I) Compounds. 4. Syntheses of Tris(tripheny1 phosphite)cobalt(I) Halides LAWRENCE W. GOSSER AIC605 1OB
Received July 14, 1976
Three methods of preparation of C O X [ ( P ~ O ) ~(XP = ] ~C1, Br) are given. Conversion to the known cobalt(1) compounds (Co[(EtO)3P]5)+CI-and CoBr (CO) [P(OPh)3] is demonstrated. Scheme I. Syntheses of CoX[(PhO),P],(L = (PhO),P; X = C1, Br) The preparation of the four-coordinate cobalt(1) halide complexes, CoXL,, has been previously described in the cases r-I CoH(CH,CN)L, L,CoC,H,OP(OPh), where L is trimethylphosphine, triphenylphosphine, and triethyl phosphite.’ The analogous triphenyl phosphite complexes, on the other hand, have not previously been isolated even though their use as catalysts with in situ formation has been disclosed.2 The preparation and characterization of the complexes CoBr(CO),L, 80 o c 110 O C CoX[(PhO),P], has now been completed. Reaction of these t excess L CoXL, excess L + complexes with triethyl phosphite gave the known cobalt(1) vacuum CoCl[(EtO),P], complexes ( C O [ ( E ~ O ) ~ P ] ~ J +Reaction X - . ~ ~ of C O B ~ [ P ( O P ~ ) ~ ] ~ CH,CN with excess carbon monoxide gave the known cobalt(1) halide complex, C O B ~ ( C O ) ~ [ ( P ~ O )2., P ] COX, + excess L The methods developed for the preparation of CoC1[(PhO),P], and CoBr[(PhO),P], are summarized in Scheme is close to the values previously observed for the analogous I. The two top routes use H X to react with a cobalt hydride complexes with tertiary phosphine* or triethyl phosphiteIb or cobalt aryl. The center two routes are ligand-exchange ligands. reactions and do not require reduction or oxidation. The Preparation from Cobalt Hydrides and HX bottom route is a reduction of cobalt halide by metallic zinc. All of the routes work fairly well but have individual diffiThe triphenylphosphinecobalt hydride complexes culties. The bottom route in particular is simple in materials CoHN2[Ph3PI3and CoH3 [Ph3P] are reported to produce the and concept, but can easily go wrong. four-coordinate complex CoCl[Ph,P], readily by reaction with The magnetic moments of the C O X [ P ( O P ~ ) ~complexes ] HC1 under mild condition^.^ Several attempts were made to were found to be about 3.2-3.4 bBat room temperature. This produce C O C ~ [ ( P ~ O )by ~ Pthe ] ~ reaction of the well-known
- -
lzn
428 Inorganic Chemistry, Vol. 16, No. 2, 1977
Lawrence W. Gosser
cobalt hydride CoH[(Ph0),PI4 with HC1. The hydride was surprisingly unreactive in HCl solutions under mild conditions. Experiments at temperatures from room temperature up to about 100 OC showed that CoH[P(OPh),I4 could be destroyed by anhydrous HCl, and in some cases dark green solutions were obtained. In none of these cases, however, was CoC1[(PhO),PI3 detected. The visible spectra suggested that the green colors were produced by a strong tail out of the UV region combined with a typical cobalt(I1) pattern in the 550-700-nm region. Recently, much more reactive cobalt complexes with triphenyl phosphite ligands became a ~ a i l a b l e . ~Both CoH-
A number of trial reductions of CoClZ/(Ph0),P with zinc in nitrile solvents produced red-amber liquids and tan-gray solids which were shown by 'H NMR to be impure CoH[ (Ph0)3P]4. However, with increasingly careful control of the reaction conditions it was possible to prepare CoX[P(OPh)J3 by zinc reduction of solutions prepared from Coxz and (Ph0)3P. Reaction with Carbon Monoxide A benzene solution of CoBr [ (Ph0)3P] readily absorbed carbon monoxide even below 1 atm at room temperature yielding a yellow crystalline solid which was shown (see Experimental Section) to be the same carbonyl complex, (CH,CN) [ P(OPh),] and [ (PhO),P] , C O C , H ~ O P ( O P ~ ) ~ C O B ~ ( C O (Ph0)3P]2, )~[ as prepared by Hieber and Dureacted with anhydrous HCl or HBr, and the complexes c h a t ~ c h .It~ was also shown that C O B ~ [ ( P ~ O )could ~ P ] ~be CoX[P(OPh),], were formed. In a set of preliminary exregenerated from C O B ~ ( C O ) ~ [ ( P ~ Oin) ~ P ]presence ~ the of periments analysis of the gas phase by mass spectroscopy excess (Ph0)3P.7b showed that the reaction between CoH(CH,CN) [ (PhO),P], Experimental Section and HCl liberated hydrogen gas in an approximately stoiThe IR spectra were recorded with a Perkin-Elmer Model 467 chiometric quantity when either 1 equiv or excess HC1 was spectrometer. A Beckman Model 25 spectrometer and a Cary Model introduced. This reaction appeared to be complete in a few 17 spectrometer were used for the electronic spectra. The 'H N M R minutes at room temperature. The reaction of HC1 with spectra were obtained with a Varian A-60 instrument. The 31pN M R
,
-
-
[(PhO),P] 3CoC6H40P(OPh)zwas not as rapid. These mixtures were typically left at room temperature for several hours before the product, CoCl[ (PhO),P],, was isolated. Presumably triphenyl phosphite was the other product of this reaction, although this was not confirmed experimentally. Preparation from CoCI[ (Et0)3P]3 Heating a mixture of excess (PhO),P and CoCl[(EtO),P], to just under the decomposition point (ca. 110 "C) under vacuum gave smooth conversion of the brown triethyl phosphite complex to the green triphenyl phosphite complex. There was no black deposit or other indication of decomposition during this transformation. The reverse reaction, conversion of CoCl[(PhO),P] to (Co[(EtO),P],)+Cl-, takes place readily at room temperature. This reaction of CoCl[(PhO),P], with (EtO),P in acetone or acetonitrile gave a yellow color quickly and a 390-nm peak appeared in the visible spectrum. Subsequent experiments with CoBr[(PhO),P], showed that the rate of the reaction with (EtO),P was greatly influenced by solvent polarity. The reaction was faster in acetone than in T H F or benzene. In 50 vol % THF in acetone semiquantitativeexperiments showed that the rate of appearance of the -390-nm peak of (Co[ (Et0)3P]$Br- was strongiy dependent on the concentration of (EtO),P. Similar observations were made in the reaction of CoCl[(EtO),P], with excess (EtO),P.lb In one case the {Co[(EtO),P] J+Cl- product of this reaction was isolated and characterized as described in the Experimental Section. There was no indication of reaction of CoCl[(PhO),P], with excess (PhO),P to produce five-coordinate complexes even in neat (PhO),P. This may be due to the large size of the (Ph0)3P ligand. The size of the ligands has been shown to be important in nickel(0) chemistryS6 Reduction of CoC12/(Ph0)3P Mixtures The reduction of CoC12/(Ph0),P mixtures can give a variety of results depending on the details. Under some circumstances, especially with sodium borohydride and hydrated cobalt salts such as C O ( N O , ) ~ . ~ H ~aOgood , yield of the hydride CoH[(PhO),PI4 can be obtained from ethanol solution7aor from acetonitrile solution.2b Similar conditions have been employed to prepare the CoXL3 cobalt(1) compounds from CoX2/L solutions in ethanol by reduction with sodium borohydride or metallic zinc in cases where the ligands are triarylphosphines or diarylalkylphosphines.8 The catalytic behavior of CoX2/(PhO),P/reducing agent mixtures suggested the formation of CoX[(PhO),P], compounds but these were not isolated.'
-
spectra were obtained with a Brucker 90-MHz instrument modified for Fourier transform operation. The melting points were obtained with an uncalibrated Me1 Temp apparatus. Commercial anhydrous cobalt salts were used. The triphenyl phosphite was a commercial sample. The tetrahydrofuran (THF) was distilled from lithium aluminum hydride and stored over sodium in nitrogen. Commercial anhydrous acetonitrile was used. The reaction mixtures were protected from contact with air. A nitrogen-filled glovebox was used for many of the operations. The magnetic moments were determined by the Evans method' using a solution of 9.4 ml of benzene, 0.20 ml of triphenyl phosphite, and 0.30 ml of tetramethylsilane as the solvent. Preparation of C O C ~ ( P ~ O ) ~from P ] , COH(CH~CN)[(P~O)~P]~ and HCI. A solution of 10 g (9.7 mmol) of C O H ( C H , C N ) [ P ( O P ~ ) , ] , ~ ~ in 100 ml of T H F was stirred at room temperature while 35.5 ml of an anhydrous 0.335 N solution of HC1 in T H F was added at ca. 5 ml/min. When the HCI addition was complete, the solution was stirred for an additional 5 min at room temperature. The pressure in the flask was then reduced and the volatiles were evaporated without warming the flask. The green solid residue was dissolved in benzene at room temperature. The solution was filtered and the volatiles were removed under vacuum. The residue was redissolved in benzene and the solution was filtered. The green solid product was precipitated by the addition of pentane. This was collected and dried; yield 8.5 g. A repeat of this procedure gave 8.3 g of green solid. The combined product (ca. 17 g) was dissolved in 40 ml of benzene, and the solution was filtered. Hexane (ca. 70 ml total) was added in increments over several hours giving 11 g (14 mmol, 70% yield) of large dark green crystals which were dried under vacuum at room temperature; mp ca. 111 "C dec (evacuated capillary). Anal. Calcd for C, 63.26; H, 4.42; Co, 5.75; CI, 3.46. Found: C, C54H4509P3C~Cl: 63.12; H , 4.49; Co, 5.60; CI, 4.03. Visible spectrum (in benzene): 758 nm, t 254; 710 nm (sh), t 200; 413 nm, t 106. The magnetic moment was 3.4 p B at 20 "C.
-
-
Preparation of COCI[(P~O)~P]~ from [ (Ph0)3P]3CoC6H40P(OPh)* and HCI. A solution was prepared from 0.65 g (0.5 mmol) of [(Ph0)3P]3CoC6H40P(OPh)p and 5.0 ml of THF, and 2.5 ml of anhydrous HCI/THF (0.32 N) was added. Since green color developed only slowly, the solution was left at room temperature for 4 h. The volatiles were then evaporated from the dark green solution at room temperature under reduced pressure. The residue was triturated with hexane and the hexane was evaporated under vacuum. The residue was then dissolved in ca. 5 ml of benzene and the solution was filtered. The benzene was evaporated under vacuum and the residue was triturated with hexane. The solid was collected, giving 0.35 g (0.45 mmol, 90% yield); mp ca. 11 1 "C dec (evacuated capillary). The magnetic moment was 3.3 y g at 43 "C. Visible spectrum: 758 nm, t 230; 710 nm (sh), t 181; 413 nm, t 85. Preparation of COC~[(P~O),P]~ from COCI[(C~H~O)~P]~. A mixture of 3 ml of triphenyl phosphite and 1.2 g (2.0 mmol) of CoC1[(EtO)3P],'bwas placed in a small flask with a magnetic stirring bar.
Tris(tripheny1 phosphite)cobalt(I) Halides
Inorganic Chemistry, Vol. 16, No. 2, 1977 429
The flask was attached to a vacuum system and the pressure was H, 4.30; Br, 7.20. Visible spectrum: e 90 at 425 nm; 715 (sh) nm; reduced to ca. 0.1 mmHg. The mixture was stirred while the e 280 at 773 nm. temperature was slowly increased by warming in an oil bath. After Preparation of C O B ~ ( C O ) ~ [ ( P ~ Ofrom ) ~ P ]C~O B ~ [ ( P ~ O ) ~ PA] , . 70 min with the oil bath at ca. 110 OC the flask had lost 0.85 g of flask was charged with 5.35 g (5.0 mmol) of CoBr[(Ph0),Pl3 and volatile material which was collected in the cold trap. There was no 100 ml of xylene. It was evacuated with an aspirator and filled with sign of decomposition to black magnetic solid in the pot. The clear carbon monoxide to 0 psig (1 atm). When the magnetic stirrer was green oily residue was left at room temperature under ca. 40 ml of started, the pressure began to drop. As the pressure dropped, C O hexane. The dark green crystals which formed were collected, washed, was added from a calibrated stainless steel reservoir. A total of 212 and dried, giving 1.5 g (1.5 mmol, 73% yield). Examination of the psi drop from a 19.9-ml volume of C O was required to maintain the IR and 'H N M R spectra of the volatiles recovered from the cold trap flask at 1 atm. This corresponds to ca. 2 mol of CO/mol of cobalt. suggested that they were primarily triethyl phosphite. The green solid The reaction mixture was cooled to -35 OC for 2 h before filtration. was crystallized three times from benzene/hexane to give 0.30 g (0.3 The yellow solid was washed with pentane and dried at room temmmol, 15% yield) of dark green shiny crystals, mp ca. 11 1 OC dec perature and 10 p pressure for 3 h (4.0 g, 4.9 mmol, 98% yield). The (evacuated capillary). The magnetic moment was perf = 3.3 p~ at 'H N M R showed that the material contained ca. one xylene per six 43 OC. Visible spectrum: 758 nm, c 238; 710 nm (sh), t 189; 413 (PhO)3P units and that it was diamagnetic. Anal. Calcd for nm, e 101. The Et-0 and Ph-0 groups have quite different IR spectra, [(C6H50)3P]2CoBr(CO)~1/3CsHlo: Co, 6.93; C, 57.40; H, 3.95; Br, especially in the 2700-3500-cm-' region. The IR spectra of hexa9.39. Calcd for [(C6H50)3P]2CoBr(CO)2:Co, 7.23; C, 55.96; H, chlorobutadiene mulls showed that there could not be more than a 3.71; Br, 9.8. Found: Co, 7.28; C, 57.36; H, 4.17; Br, 9.98. few percent of C2H5O in the final solid product. Material from a similar, smaller scale preparation turned green Preparation of C O B ~ [ ( P ~ O ) ~from P ] , CoH(CH,CN)[P(OPh),], and and melted at ca. 105 OC (evacuated capillary). The filtrate yielded HBr. A solution of 5.0 g (4.8 mmol) of C O H ( C H ~ C N ) [ P ( O P ~ ) ~ ] ~an ~ ~oil (about 30% of starting material weight) which was shown by in 50 ml of T H F was stirred at room temperature and 17 ml of a 3'P and IH N M R to be predominantly triphenyl phosphite. The ,'P freshly prepared 0.31 N solution of anhydrous HBr in ether was added. N M R spectrum of the dicarbonyl complex consisted of a single peak After an additional 3.5 min the pressure was reduced and the volatiles 147 ppm downfield from external 85% H3P04. This remained a single were evaporated. The residue was triturated with hexane and the peak from room temperature down to -100 OC. The addition of hexane was removed under vacuum. The residue was then recrystriphenyl phosphite at room temperature resulted only in the appearance of the sharp 127-ppm peak of free triphenyl phosphite. tallized three times from ca. 20 ml of benzene by adding ca. 100 ml Addition of excess C O to a solution of C O B ~ [ ( P ~ O ) ~produced P], a of hexane to give 2.4 g (2.2 mmol, 47% yield) of dark green crystals, solution with the same 31PN M R spectrum as given by a mixture of mp ca. 124 OC dec (evacuated capillary). Anal. Calcd for C54H4509P3C~Br: C, 60.63; H, 4.24; Co, 5.51; Br, 7.47. Found: C, the yellow solid and (Ph0)3P. This confirmed that the addition of C O was accompanied by the loss of triphenyl phosphite from the 60.39; H, 4.33; Co, 5.72; Br, 7.74. Visible spectrum (benzene): 773 nm, e 257; 425 nm, t 78. The magnetic moment was perf = 3.2 p~g complex. The IR (Nujol) showed two carbonyl bands 2030 and 1950 a t 37 OC. cm-' in ca. 1:2 intensity ratio. Preparation of CoCI[(Ph0),PI3 by Zinc Reduction of CoCI2 in Preparation of (Co[(EtO)3P]5J+CIfrom C a ( P h 0 ) , P l 3 . A solution CH3CN. A solution of 7.8 g (60 mmol) of commercial anhydrous of 250 mg (6.24 mmol) of C O C ~ [ ( P ~ O )and ~ P ]0.50 ~ ml of triethyl cobalt dichloride was dissolved in 300 ml of commercial anhydrous phosphite in 2.5 ml of acetone was left at room temperature for 6 acetonitrile. The solution was cooled to 15 OC and 3.0 ml of water h. The volatiles were removed under vacuum and the residue was was added followed by 90 ml (342 mmol) of triphenyl phosphite. After triturated with hexane. The yellow solid was collected, dried, and thorough mixing 1.80 g (28 mmol) of zinc dust (Fisher, Tech) was weighed, giving 150 mg (100 % yield). The 'H N M R spectrum and added followed by immediate vigorous shaking of the reaction vessel the IR (hexachlorobutadiene mull) were the same as those of authentic to keep the zinc suspended. After 5 min of vigorous agitation, the ( C O [ ( E ~ O ) ~ P ] ~ ) " The C ~ - visible . spectrum was also the same as that mixture was left at room temperature (20 "C) for 1 h and then filtered. of ( C O [ ( E ~ O ) ~ P ] ~ J "Le., C ~ - a, single peak t ca. lo' at 388 nm.lb After ca. 24 h at about 15 OC the supernatant green liquid was Registry No. C O C I [ ( P ~ O ) ~ P ]61202-74-2; ,, CoBr[(PhO),P],, decanted from the large dark green crystals which had formed in the 61026-10-6; C O B ~ ( C O ) ~ [ ( P ~ O ) ~14653-41-9; P],, (CO[(E~O)~P]~)+C~-, filtrate. These were washed with pentane and dried, giving 6.1 g (6 53701-72-7; CoH(CH3CN) [P(OPh),l3, 58527-60-9; mmol, 1OOh yield) (4.7 g in another case). A repeat of this procedure I [(P~O)~P]~COC~H~~P(OP~)~, 54870-2 1-2; COCI[(EtO)3P] 3 , with twice these amounts gave 19.3 g (19 mmol, 16% yield) of large 15488-43-4. green crystals. Thirty grams of material prepared in this way was dissolved in 200 ml of benzene/lO ml of (PhO),P at room temperature. References and Notes Slow dilution of the filtered solution with 700 ml of hexane gave large (a) H. F. Klein and H. H. Karsch, Inorg. Chem., 14,473 (1975); (b) green crystals which were similarly recrystallized from 120 ml of L. W. Gosser and G. W. Parshall, ibid., 13, 1947 (1974), and references benzene/6 ml of triphenyl phosphite/500 ml of hexane. The large therein. green crystals weighed 24.1 g. Anal. Calcd for C54H4509P3CoCl: W. C. Drinkard, Jr., U S . Patent 3655 723 (1972); W. C. Drinkard, Jr., C, 63.26; H, 4.42; C1, 3.46. Found: C, 62.82; H, 4.49; C1,3.54. Visible and B. W. Taylor, US.Patents 3 775461 (1973) and 3 579 560 (1971). W. Hieber and H. Duchatsch, Chem. Ber., 98, 2530 (1965). spectrum: 410 nm, t 84; 715 nm (sh); 755 nm, t 210. R. C. White'and J. G. Thatcher, U S . Patent 3692 864 (1972); A. Sacco Preparation of CoBr[(Ph0)3P]3 by Zinc Reduction of CoBr2 in and M. Rossi, Inorg. Chim. Acta, 2, 127 (1968). Acetonitrile. A solution of 8.8 g (40 mmol) of commercial anhydrous L. W. Gosser, Inorg. Chem., 14, 1453 (1975); 15, 1348 (1976); U S . CoBr2 in 200 ml of commercial anhydrous acetonitrile was cooled Patent, 3927056 (1975). to 20 OC and 60 ml (228 mmol) of triphenyl phosphite was added. L. W. Gosser and C.A. Tolman, Inorg. Chem., 9, 2350 (1970); C. A. Tolman, W. C. Seidel, and L. W. Gosser, J . Am. Chem. Soc., 96, 53 Zinc dust (1.25 g, 19 mmol) was then added and the mixture was ( 1974). immediately shaken vigorously for 5 min. The mixture was then left (a) J. J. Levison and S. D. Robinson, J . Chem. SOC.A , 96 (1970); (b) 15 min at room temperature before the green liquid was decanted. L. W. Gosser, Inorg. Chem., following paper in this issue. The green liquid was then left 3 days at 20 OC, and 26 g (24 mmol, M. Aresta, M. Rossi, and A. Sacco, Inorg. Chim. Acta, 3, 227 (1969); 60% yield) of large green crystals was collected. Anal. Calcd for L. Sacconiand S. Midollini, J. Chem. Soc., Dalton Tram., 1213 (1972). C54H4509P3C~Br: C, 60.63; H, 4.24; Br, 7.47. Found: C, 60.80; D. F. Evans, J. Chem. SOC.,2003 (1959).